This pattern of Chinese vs. Arabic numerals processing in the foveal and parafoveal conditions has been repeatedly found in previous behavioral studies using similar paradigms [55], [57].
The second conclusion is that the P2p ERP distance effect is common to Arabic numerals and non-symbolic dot arrays, suggesting that it does indeed reflect an abstract and notation-independent neural activation that is invoked for both symbolic and non-symbolic numerical processing.
One possible interpretation of these differences is that small Chinese numerals evoked subitizing processing as small concrete quantities when lacking attentional resources.
No matter subitizing is pre-attentive or not, when lacking attentional resources, multi-representational numerals might evoke dissimilar subitizing processing than single-representational numerals and thus demonstrate observable difference than single-representational numerals.
Eventually, there can be some more input resources to process more complicated tasks (e.g. processing of numerals, physical units, degrees and titles, and special symbols like @, &, or %).
These behavioral results replicated our previous findings that, when presented at the parafoveal location, Chinese and Arabic numerals demonstrate different number processing effects depending on the magnitude [55], [57].
However, participants' performance differed greatly for these two visually and semantically similar numerals (Fig. 2), further indicated that it is the common properties of small Chinese numerals as a whole (e.g., being multi-representational numerals), rather than the individual characteristic of each numeral per se, that contributes to their processing advantage over small Arabic numerals.
As a concrete quantity, the Roman and Chinese numerals for one, two and three could potentially evoke subitizing processing as several separate lines and thus facilitate the access of magnitude information in these numerals.
More specifically, could properties that adhere to a concrete quantity facilitate the processing of these multi-representational numerals to some extent?
In our results, Chinese and Arabic numerals showed no difference in behavioral results when being presented at the foveal location, which is in line with previous studies of Chinese [15], [17], [18] and Japanese Kanji number processing [16], [19], [61], indicating that Chinese numerals are as easy to be identified as Arabic numerals [15], [18].
